See discussions, stats, and author profiles for this publication at: [Link]

net/publication/342894453

Experimental investigation of impact of long rod eroding projectiles against

rolled homogeneous armour

Article in Proceedings of the Institution of Mechanical Engineers Part L Journal of Materials Design and Applications · July 2020
DOI: 10.1177/1464420720940766

CITATIONS READS

3 526

3 authors:

Sana Zaki Emad Uddin

University College Dublin National University of Sciences and Technology
12 PUBLICATIONS 98 CITATIONS 105 PUBLICATIONS 1,171 CITATIONS

SEE PROFILE SEE PROFILE

Aamir Mubashar
National University of Sciences and Technology
80 PUBLICATIONS 1,179 CITATIONS

SEE PROFILE

All content following this page was uploaded by Sana Zaki on 28 April 2021.

The user has requested enhancement of the downloaded file.

Original article

Proc IMechE Part L:

J Materials: Design and Applications
Experimental investigation of impact of 2020, Vol. 234(10) 1335–1342
! IMechE 2020
long rod eroding projectiles against rolled Article reuse guidelines:
[Link]/journals-permissions
homogeneous armour DOI: 10.1177/1464420720940766
[Link]/home/pil

Sana Zaki1, Emad Uddin2 and Aamir Mubashar2

Abstract
Long rod eroding penetrators exhibit hydrodynamic behaviour upon impact against the target plates. An experimental
study was carried out to investigate the factors affecting the penetration of tungsten alloy rods against rolled homog-
enous armour at ordnance velocities. Experiments were carried out for range of projectile length to diameter ratios to
determine the penetration efficiency of the projectile, where the projectile penetration efficiency is the penetration
depth per unit projectile length. Analytical Alekseevski–Tate model was implemented to predict the penetration depth
across the length to diameter range and the results were compared with the experimental data. The tungsten percent-
age of the projectile material was also varied to change its mechanical properties and thus observe the effect of the same
on the penetration depth. The experimental results showed a linear relationship between the penetration efficiency and
the length to diameter ratio across the considered range of length to diameter ratios. The Alekseevski–Tate model
predictions were not consistent across as the model does not consider the effect of length to diameter ratio. The
projectiles with high impact energy were able to perforate the target plate.

Keywords
High velocity impact, eroding projectiles, analytical modelling
Date received: 18 May 2020; accepted: 18 June 2020

Introduction
target was found to increase with increase of L/D
The performance of long rod projectiles at high ratio of the projectile.2,10,11 It has also been experi-
impact velocities (1–5 km/s) has remained an area mentally observed that there is a linear relationship
of research interest. Deformation at the target inter- between the projectile velocity, v, and the penetration
face is hydrodynamic in nature. The impact response velocity, u, which is of general form u ¼ a þ bv where
of the long rods is dependent upon the range of a and b are constants. Senthil et al.12 reported the
impact velocity.1 At lower impact velocities, the experimental ballistic performance of scaled tungsten
long rods behave like rigid projectiles and at higher heavy alloy (WHA) against steel targets and found
velocities the projectiles erode during the impact. A that penetration velocity and impact velocity were
transition region where the projectile acts as a varying linearly for a velocity range of 900–1400 m/s.
deformable rod is observed between the low and Orphal and Anderson13 proposed that this linear rela-
high impact velocities. The penetration performance tionship is fundamental in nature as it is applicable
of the projectiles depends upon many factors includ- for a wide range of projectile/target combinations
ing impact velocity, nose shape, angle of attack, mate-
rial properties, and length to diameter (L/D) ratio.
One measure to determine the penetration perfor- 1
Wah Industries, Wah, Pakistan
mance of the projectile is projectile penetration effi- 2
Computational Mechanics Group, Department of Mechanical
ciency that is the penetration depth per unit projectile Engineering, School of Mechanical and Manufacturing Engineering,
length (P/L). Effects of impact velocity and L/D were National University of Sciences and Technology (NUST), Islamabad,
Pakistan
investigated by many researchers.2–6 Experimental
Corresponding author:
observations have shown that eroding long-rod pen- Aamir Mubashar, National University of Sciences and Technology, H-12,
etration is a linear function of impact velocity5–9 over Islamabad 44000, Pakistan.
a range of 1–8 km/s. Penetration of the projectile in Email: aamir@[Link]
1336 Proc IMechE Part L: J Materials: Design and Applications 234(10)

and a and b are related to materials of projectile and This research is based on experimental study to
target. Penetration performance of long rod projec- determine the penetration performance of tungsten
tiles is also dependent on shape of penetrators such as rods against rolled homogenous armour (RHA)
ogive, spherical, conical and flat.2,10,14–17 Rosenberg plates. Two sets of experiments were independently
and Dekel18 observed that a critical velocity separates carried out where the L/D ratio of the projectile was
the deceleration behaviour of rigid long rod projec- varied in different ranges. Many of the experimental
tiles where the deceleration is linear below the critical results in the literature provide data for L/D ratios of
velocity. The critical velocity is dependent upon the up to 15 and data for higher L/D is available for
target strength and the shape of the projectile. A rela- specific ratios. This study provides a set of experimen-
tionship for the projectile penetration was also sug- tal data for L/D ratios of 15.6 to 25.2 and thus
gested based on the experimental observations, which explores a relatively wide range of L/D with multiple
showed that penetration depth is dependent upon the points in between the two chosen limits. The tests
densities of the projectile and the target, initial veloc- were performed on the full-scale projectiles and the
ity of the projectile, critical velocity and nose shape of depth of penetration was recorded for each case. The
the projectile. Alekseevski–Tate analytical model, that is commonly
The projectile and target impact is a complex phe- used for long rod eroding projectiles was used to pre-
nomenon and several analytical and numerical dict the penetration depths and was validated by com-
models have been suggested to simulate and predict parison with the experimental data. This shows the
aspects of the impact event. Although finite element- validity of the model across a broader range of L/D
based numerical models provide detailed information ratios and emphasizes many of its short comings
about various aspects of the impact event by success- reported earlier in the literature. Finally, an indepen-
fully simulating all the phases of high velocity impact, dent set of experiments explored the three different
these models require expertise to setup, computation- percentages of tungsten (W) in the projectiles of var-
al power to execute and are time consuming. In con- ious lengths and its impact on the penetration depth.
trast, the analytical models can be implemented
relatively easily, provide results in less time and
good estimates of several aspects of the impact phe-
Experimental setup
nomenon are achieved. Allen and Rogers19 based An experimental programme was carried out to study
their analytical model on the experimental observa- the geometric and material changes affecting the pen-
tions of penetration of metal rods in a semi-infinite etration in the target plate. The projectile was made of
target. Good predictions of the experimental penetra- Tungsten alloy W93 with small proportions of W-Ni-
tion depths were provided by the model. The model Fe and had density of 17,680 kg/m3. During the first
also showed that at higher velocities for a projectile/ set of experiments, the projectile’s L/D ratio was
target combination, a hydrodynamic limit is achieved changed from 16.1 to 18.6 by changing the length of
qffiffiffiffiffiffiffiffiffiffiffiffiefficiency.
that inhibits further increase in penetration ffi the projectile only and keeping its diameter fixed. In
The hydrodynamic limit is given by qp =qT where qp the second set of the experiments, both the diameter
and qT are densities of projectile and target materials, and the length of the projectile were changed to
respectively. One of the commonly used analytical achieve L/D ratios between 15.6 and 25.2. The third
models was proposed separately by Alekseevski20 set of experiments was carried out where length as
and Tate21 and is referred to as Alekseevski–Tate well as the tungsten percentage in the projectile was
(AT) analytical model. The AT model is based on a varied. The design of experiments is shown in Table 1.
modified form of Bernoulli equation and provides Sketch of the experimental setup is shown in Figure 1
predictions based on the steady-state phase of the where two chronometers were installed to measure
projectile penetration. Effect of target and the projec- the velocity of the projectiles: one at the launching
tile strengths has been included in the model defini- end at 50 m from launch point and the other at
tion. The model equations can be solved numerically target end at 400 m from the launch point. In addi-
to obtain time-dependent penetration and projectile tion, a high-speed camera was also placed to capture
velocities, P/L and penetration of the projectile. A the impact of the long rod projectiles with the target
few closed form approximate solutions have been plate. The target plate was made of RHA and was
proposed for the AT model22–24 which provide good placed at 500 m from the launch point of the projec-
results when compared to the numerical solution. tile. The target plate was 1000  1000  562 mm in
Frank and Zook25 recommended a modification to size and was inclined at an angle of 30 to increase
the AT model to incorporate multiple phases of pen- the effective thickness which resulted in an effective
etration in the model equations. Galanov et al.26 length of approximately 650 mm. Target plate density
modified the AT model to make it applicable for was 7830 kg/m3 and it had hardness of 235 BHN. The
elastic-plastic as well as brittle materials such as mechanical properties of the projectile and the target
ceramics. An analytical model for impact of long plate are given in Table 2. For each experimental data
rod steel projectiles on concrete targets has been pro- point, 5 projectiles were impacted on the target
posed by Liu et al.27 plate and an average velocity of 1635 m/s was
Zaki et al. 1337

Table 1. Design of experiments.

S No Parameter Values

1 Projectile length, L (mm) with fixed diameter 450, 460, 470, 480, 490, 500, 520
2 Ratio of projectile length to diameter, L/D with 15.6, 16.5, 17.8, 18.5, 19, 20.5, 21, 22,
varying lengths and diameters 23.4, 24.1, 25.2
3 %age of tungsten (W) in the projectile material 91%, 92%, 93%

Figure 1. Schematic representation of the experimental setup.

Table 2. Mechanical properties of the projectile and target Table 3. Experimental penetration depths after impact testing
materials. with varying projectile lengths for the first batch.

Parameter Symbol Unit Projectile RHA Length of Depth of

projectile, penetration,
Density q kg/m 3
17,680 7830 S No L (mm) P (mm) P/L
Shear modulus G GPa 134 75
1 450 461 1.024
2 460 470 1.022
measured based on all the experimental data sets. 3 470 477 1.015
Zero angle of impact between the rod and the target 4 480 482.2 1.005
plate was assumed based on the high-speed camera 5 490 491.5 1.003
footage. After each experiment, the penetration and 6 500 508.2 1.016
7 520 513 0.987
distal sides of the target plate were examined.

Experimental results and discussion

Figure 2. A linear increase in the penetration depth
In the first experimental set, the length of the projec- of the projectile was observed with increasing P/L
tile was progressively increased from 450 to 520 mm where a linear curve fit is shown by the dotted line.
where the length was increased from 450 to 500 mm Ratio of change in penetration depth to P/L with the
with increment of 10 mm and was further increased change in L/D is plotted in Figure 3. An increase in
with an interval of 20 to 520 mm. The diameter of the L/D ratio decreases the projectile penetration efficien-
projectile was kept fixed at 28 mm which provided a cy which is consistent with the established literature
L/D ratio range of 16.1 to 18.6. No dimensional scal- that the P/L for a long eroding rod is lower than the
ing was carried out and the projectiles used in the same for a short eroding rod projectile. Figure 3
actual system were tested. The depth of penetration shows that this decrease is linear for the plotted
for each P/L is given in Table 3 and is plotted in L/D range. Another set of experiments was conducted
1338 Proc IMechE Part L: J Materials: Design and Applications 234(10)

Figure 2. Variation in depth of penetration with change in Figure 4. Variation in depth of penetration with change in L/D.
projectile length.

Figure 3. Change in P/L with increasing L/D for eroding Figure 5. Change in P/L with increasing L/D for eroding long
long rods. rods for greater range of L/D ratios.

first set of experiments. For this larger range of L/D

Table 4. Experimental penetration depths after impact testing
with varying L/D for second batch. ratios, the penetration depth increased linearly as can
be seen in Figure 4. This is expected for any kinetic
Length of Diameter of Depth of energy projectile. The linear relationship between the
projectile, projectile, penetration, L/D and P/L ratios also exists for this higher range of
S No L (mm) D (mm) L/D P (mm)
L/D ratios where projectile penetration efficiency
1 437 28.01 15.6 460 decreased linearly with increasing L/D as shown in
2 460 27.88 16.5 470 Figure 5. A decrease of 15.4% in P/L ratio was
3 480 26.97 17.8 482 observed for an increase of 38.1% in L/D ratio of
4 500 27.03 18.5 491 the projectile. When the results of the two separate
5 520 27.37 19 500.5 sets of the experiments were plotted together in the
6 540 26.34 20.5 508.1
same graph, Figure 6, the two sets of data showed
7 560 26.67 21 520
8 580 26.36 22 527.3 that they follow the same linear trend line with
9 600 25.64 23.4 542 good overlap. Although, the L/D ratios were achieved
10 620 25.73 24.1 551 in the first set by changing the length and keeping the
11 640 25.40 25.2 570 diameter fixed and in the second set by changing both
the length and the diameter, the penetration efficiency
and L/D ratio have the same relationship. This shows
to further explore the effect of L/D ratio on penetra- that for the projectile–target combination at the
tion depth where an L/D ratio of up to 25.2 was inves- chosen velocity and L/D ratios, an overall relation-
tigated. The L/D ratios given in Table 4 were achieved ship between the L/D and projectile penetration effi-
by changing both the length and the diameter of the ciency exists that can be used to determine the
projectile. The diameter of the projectile was penetration efficiency of any L/D. All the experiments
decreased gradually from the 28 mm used in the conducted were below the hydrodynamic limit of the
Zaki et al. 1339

Figure 6. Combined results of the two sets of experimental Figure 7. Penetration depth of the projectile as predicted by
data for change in P/L with increasing L/D. the Alekseevski–Tate (AT) analytical model.

chosen target–plate combination that is calculated by the experimental data. The penetration depth pre-
qffiffiffiffiffiffiffiffiffiffiffiffiffi
using qp =qT where qp and qT are densities of pro- dicted by the AT model and the experimental pene-
tration data are shown in Figure 7 for dynamic flow
jectile and target materials, respectively. For the stress of projectile material, Yp ¼ 1 GPa11 and pene-
selected projectile–target system, the hydrodynamic tration resistance of target material, RT ¼ 4 GPa.11
limit is 1.5. Thus, no stagnation in the penetration The chosen value of dynamic flow stress has been
depth was observed during the experiments. used in literature for WHA projectiles11 while a
The experimental penetration depths were com- range of RT ¼ 3–5 GPa11 is given for steel targets.
pared with the Alekseevski–Tate (AT) analytical The AT model predicted a linear trend of increasing
model that was separately presented by penetration depth with increasing L/D; however, the
Alekseevski20 and Tate.21 The AT model is used to slope of the linear curve fit to the predicted values is
predict the behaviour of eroding long rod projectiles. higher than the same for the experimental values.
The governing equations of the AT model are given in Thus, the AT model under predicted the penetration
equations (1) to (4) depth at the start of the experimental range and over
predicted the penetration depth at the higher L/D
1 1
q ðv  uÞ2 þ Yp ¼ qT u2 þ RT (1) ratios. The intersection point of the experimental
2 p 2 values trend line and the AT model trend line can
be moved by changing the values of Yp and RT as
dv consensus on selection of these values does not exist
qp l ¼ Yp (2)
dt in literature.28 Thus, the model can be over or under
predicting the experimental results based on the initial
dl values of Yp and RT . AT model is also insensitive to
¼uv (3)
dt change in L/D and thus its effect on the prediction of
penetration depth. However, as the results of the
dp experiments show, the P/L is changing with changing
¼u (4)
dt L/D ratio and thus AT model is not able to predict
the penetration depth accurately.11 The correspon-
where qp and qT are densities of projectile and target dence between the numerical solution of the AT
materials, respectively, Yp is the dynamic flow stress model and its approximate solution remained lower
of projectile material, RT is penetration resistance of than 1.3% for the considered range of L/D. The
the target material, u is the penetration velocity, v is approximate solution consistently predicted higher
the projectile velocity, l is the P/L and p is the pro- penetration depth than the numerical solution.
jectile penetration. The model provides variation of However, the AT model provides quantitative esti-
u; v; l and p based on the impact time t. The model is mates of projectile and impact velocities as well as
to be solved numerically to determine the time vary- the length of the eroding projectile during the
ing variables. One of the approximate solutions for impact. The predicted velocities and the eroding
the AT model was presented by Jiao and Chen24 that length for a projectile of initial length of 640 mm
provides equations that can be directly solved to and L/D ¼ 25.2 are shown in Figures 8 and 9, respec-
obtain the of u; v; l and p. Both the numerical tively. The AT model predicts a sharp decrease in the
approach and the approximate solution of Jiao and projectile and impact velocities towards the end of the
Chen24 were used to predict the penetration depth for penetration phase and that the projectile would
1340 Proc IMechE Part L: J Materials: Design and Applications 234(10)

Table 5. Effect of change of tungsten (W) %age on projectile

penetration in 600 mm plate.
Projectile Impact
S length UTS Weight energy Penetration in
No Material (mm) (MPa) (kg) (J) 600 mm target

1 93% W 550 1160 4.2 89 Penetrated target

2 600 1150 4.6 95 Penetrated target
3 650 1150 4.9 120 Penetrated target
4 92% W 550 1469 3.8 87 Penetrated target
5 600 1469 4.0 97 Penetrated target
6 650 1469 4.0 94 Penetrated target
7 91% W 550 900 4.0 94 Penetrated target
8 600 1400 4.4 118 Perforated target
9 650 1400 4.4 107 Perforated target

Figure 8. Impact and penetration velocities for the initial

projectile length of 640 mm and L/D ¼ 25.2 as predicted by the
Alekseevski–Tate model.

Figure 9. Eroding length of the projectile with initial length

640 mm and L/D ¼ 25.2 as predicted by the Alekseevski–Tate
model.

completely erode by the end of the impact. The pre-

diction of complete erosion of the projectile is true for
all combinations of L/D considered in this work. The
change in P/L with time follows a linear path.
However, small fractions of the projectile were
found after the impact during experimentation.
The trend of predictions of velocities and projectile
erosion remained similar to what is shown in
Figures 8 and 9 throughout the considered range of
L/D which is due to the independence of the AT
model from L/D. Figure 10. Perforation of 600 mm projectile in the target
The final set of experiments was conducted to plate: (a) impact side and (b) distal side.
observe the effect of change in the %age of tungsten
(W) in the projectile material. The length of the pro-
jectile was also changed from 550 to 650 mm. projectile with 91% W was also unable to perforate
Perforation of a 600 mm plate was considered by the target. However, both the 600 and 650 mm pro-
changing the %age W from 91% to 93% and the jectiles having 91% W perforated the target plates
results are given in Table 5. It was observed that the successfully. The ultimate tensile strengths (UTS)
550 and 600 mm projectiles with any %age W were and the impact energy of the projectile material, as
unable to perforate the 600 mm plate. The projectiles determined by a Charpy impact test, are also given in
penetrated the target plate and resulted in bulging at Table 5. Although, the projectiles made of 92% W
the distal end of the target plate. The 550 mm had the highest UTS, they were unable to perforate
Zaki et al. 1341

Figure 11. Perforation of 650 mm projectile in the target plate: (a) impact side and (b) distal side.

the target plate. Thus, the strength of the projectile penetration depth. Ductile type of failure was
material does not affect the penetration depth by a observed in the perforated plates.
large amount due to the deceleration of the projectile
tail by the high strength nose of the projectile.11 The
Declaration of conflicting interests
increased impact energy of the 91% W projectiles
allowed them to perforate the target material as the The author(s) declared no potential conflicts of interest with
projectile erosion was delayed. The perforated plate is respect to the research, authorship, and/or publication of
shown in Figures 10 and 11 for the 600 and 650 mm this article.
projectiles, respectively. The failure of the target plate
is due to the ductile piercing where the material of the Funding
plate was pushed away by the impacting projectile. The author(s) received no financial support for the research,
Bulging at the entry and exit side of the perforation authorship, and/or publication of this article.
was observed. This type of failure is characteristic of
ductile material failure at velocities above 1000 m/s.29
ORCID iDs
Emad Uddin [Link]
Conclusions Aamir Mubashar [Link]
This research investigated the penetration efficiency
of long rod eroding projectiles over a range of L/D References
ratios. It was observed that the penetration efficiency 1. Lou J-f, Zhang Y-g, Wang Z, et al. Long-rod penetra-
decreased with the increasing L/D as a linear function tion: the transition zone between rigid and hydrody-
over the range of L/D ¼ 15.6–25.2. The two indepen- namic penetration modes. Defence Technol 2014; 10:
dent sets of the experiments showed that a single 239–244.
2. Anderson CE, Walker JD, Bless SJ, et al. On the veloc-
curve of penetration efficiency versus L/D can be
ity dependence of the L/D effect for long-rod penetra-
used to determine the penetration efficiencies for the tors. Int J Impact Eng 1995; 17: 13–24.
wide range of L/D considered. The analytical AT 3. Baker JR. Hypervelocity crater penetration depth and
model that is solved numerically and its approximate diameter – a linear function of impact velocity? Int J
solution were able to predict the penetration depth Impact Eng 1995; 17: 25–35.
with some success for the range of L/D ratios. The 4. Anderson CE, Walker JD, Bless SJ, et al. On the LD
selection of AT model parameters and its indepen- effect for long-rod penetrators. Int J Impact Eng 1996;
dence of L/D ratios resulted in under or over predic- 18: 247–264.
tion of the penetration velocities. The approximate 5. Kadono T and Fujiwara A. Cavity and crater depth in
solution of the AT model was in good agreement hypervelocity impact. Int J Impact Eng 2005; 31:
1309–1317.
with the numerical solution of the AT model and
6. Behner T, Heine A and Wickert M. Dwell and penetra-
thus may be used as instead of the numerical solution. tion of tungsten heavy alloy long-rod penetrators
The change in %age of W and corresponding mate- impacting unconfined finite-thickness silicon carbide
rial properties showed that the penetration behaviour ceramic targets. Int J Impact Eng 2016; 95: 54–60.
of the projectiles improved with increased impact 7. Subramanian R, Bless SJ, Cazamias J, et al. Reverse
energy whereas the UTS had minimal effect on the impact experiments against tungsten rods and results
1342 Proc IMechE Part L: J Materials: Design and Applications 234(10)

for aluminum penetration between 1.5 and 4.2 km/s. Int 18. Rosenberg Z and Dekel E. The penetration of rigid long
J Impact Eng 1995; 17: 817–824. rods – revisited. Int J Impact Eng 2009; 36: 551–564.
8. Sch€afer FK, Herrwerth M, Hiermaier SJ, et al. Shape 19. Allen WA and Rogers JW. Penetration of a rod into a
effects in hypervelocity impact on semi-infinite metallic semi-infinite target. J Franklin Inst 1961; 272: 275–284.
targets. Int J Impact Eng 2001; 26: 699–711. 20. Alekseevskii VP. Penetration of a rod into a target at
9. Goloveshkin VA and Myagkov NN. Estimating param- high velocity. Combust Explosion Shock Waves 1966; 2:
eters of fragments forming at penetration of periodic 63–66.
system of strings into a semi-infinite target. Int J 21. Tate A. A theory for the deceleration of long rods after
Impact Eng 2013; 61: 13–23. impact. J Mech Phys Solids 1967; 15: 387–399.
10. Alavi Nia A, Zolfaghari M, Mahmoudi AH, et al. 22. Walters WP and Segletes SB. An exact solution of the
Analysis of resistance of concrete target against pene- long rod penetration equations. Int J Impact Eng 1991;
tration of eroding long rod projectile regarding flow 11: 225–231.
field around the projectile tip. Int J Impact Eng 2013; 23. Segletes SB and Walters WP. Extensions to the exact
57: 36–42. solution of the long-rod penetration/erosion equations.
11. Rosenberg Z and Dekel E. Terminal ballistics. Berlin:
Int J Impact Eng 2003; 28: 363–376.
Springer, 2016.
24. Jiao WJ and Chen XW. Approximate solutions of the
12. Senthil PP, Reddy PRS, Reddy TS, et al. Scaled WHA
Alekseevskii–Tate model of long-rod penetration. Acta
long rod projectile impact against an armour steel. Hum
Mech Sin 2018; 34: 334–348.
Factors Mech Eng Defense Saf 2019; 3: 14.
25. Frank K and Zook J. Energy-efficient penetration and
13. Orphal DL and Anderson CE. The dependence of pen-
perforation of targets in the hypervelocity regime. Int J
etration velocity on impact velocity. Int J Impact Eng
Impact Eng 1987; 5: 277–284.
2006; 33: 546–554.
26. Galanov BA, Ivanov SM and Kartuzov VV. On one
14. Luk VK and Forrestal MJ. Penetration into semi-
infinite reinforced-concrete targets with spherical and new modification of Alekseevskii–Tate model for non-
ogival nose projectiles. Int J Impact Eng 1987; 6: stationary penetration of long rods into targets. Int J
291–301. Impact Eng 2001; 26: 201–210.
15. Forrestal MJ, Luk VK and Brar NS. Perforation of 27. Liu C, Zhang X, Chen H, et al. Experimental and the-
aluminum armor plates with conical-nose projectiles. oretical study on steel long-rod projectile penetration
Mech Mater 1990; 10: 97–105. into concrete targets with elevated impact velocities.
16. Frew DJ, Hanchak SJ, Green ML, et al. Penetration of Int J Impact Eng 2020; 138: 103482.
concrete targets with ogive-nose steel rods. Int J Impact 28. Crouch IG. The science of armour materials. Oxford:
Eng 1998; 21: 489–497. Woodhead Publishing, 2017.
17. Silling SA and Forrestal MJ. Mass loss from abrasion 29. Carlucci DE and Jacobson SS. Ballistics – theory and
on ogive-nose steel projectiles that penetrate concrete design of guns and ammunition. Boca Raton, FL: CRC
targets. Int J Impact Eng 2007; 34: 1814–1820. Press, 2014.

View publication stats